Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
MATH 522–01 Problem Set #1 solutions 1. Let U be a nonempty set and let R be the set of all subsets of U (i.e. the power set of U ). For the two given proposed definitions of “addition” and “multiplication”, determine whether R is a ring or not; if it is not a ring, explain why, and if it is a ring, identify its identity elements. (a) A + B = A ∪ B and A · B = A ∩ B. This is not a ring, because there is no inverse under addition. To determine an identity, we note that A + 0 = A ∪ 0 = A is only universally true if 0 = ∅; however, then since for nonempty A, there is no set (−A) such that A + (−A) = A ∪ (−A) = ∅. (b) A + B = (A ∪ B) − (A ∩ B) (also known as the “symmetric difference”) and A · B = A ∩ B. This is in fact a ring: closure is easily demonstrated, and associativity is a bit tedious but straightforward (A + B + C consists of those elements lying in exactly 1 or 3 of the sets A, B, and C). The distributeive law may be derived from noting that [(A ∪ B) − (A ∩ B)] ∩ C = [(A ∩ C) ∪ (B ∩ C)] − (A ∩ B ∩ C) The additive identity is ∅, and every element of R is its own additive inverse. There is also a multiplicative identity, U itself. 2. Prove that for a commutative ring R and x ∈ R, x is a unit of R if and only if x divides every element of R. First, let us note that the statement is vacuously true if R lacks a multiplicative identity, as neither of the conditions can be met. Clearly, x can be a unit only if xy = 1 for some x, y ∈ R, which would require that 1 ∈ R. Similarly, if x divides every element of R, then in particular x divides x, so x = kx for some k ∈ R; now, for any r ∈ R, there is an ℓ ∈ R such that r = ℓx, and so r = ℓx = ℓkx = k(ℓx) = kr, making k a multiplicative identity. Henceforth, we may thus assume that R has a multiplicative identity. Let us suppose that x is a unit, so that there is a y ∈ R such that xy = 1. Then for any r ∈ R, (xy)r = 1r = r, so r = x(yr). Thus x divides r. Conversely, if we suppose that x divides every element of R, then, in particular, x divides 1, so there is a k ∈ R such that xk = 1; since k is definitionally the inverse of x, x is a unit. 3. Prove the following two statements: (a) For any ring R and a (not necessarily finite) collection R of subrings of R, their intersection ∩ S S∈R is also a subring of R. (Hint: you may find it easier to warm up by addressing just an intersection of two subrings of R) ∩ Let T = S∈R S for brevity. Then if a, b ∈ T , it follows that a, b ∈ S for every single ring S ∈ R. Since each S is a ring, ab and a + b are in S for every S ∈ R, and thus ab and a + b are in T , demonstrating closure of T under addition and multiplication. Likewise, since each S is a ring, −a ∈ S for every S ∈ R, and so −a ∈ T , demonstrating closure under additive inversion. Finally, since every S is a ring, 0 ∈ S and thus 0 ∈ T . (b) For a ring R and a subset S of R, there is a unique smallest subring ⟨S⟩ of R which contains all the elements of S (hint: choose an appropriate collection R for the above statement). Let R ∩ contain every ring which is a superset of S and a subring of R. By the previous part, T = X∈R X is a subring of R, and for every ring X which is a superset of S and a subring of R, X ⊆ T , we may assert that T , as it contains S and is contained in every ring containing S, is the smallest such ring. Page 1 of 2 due Thursday, January 22, 2015 MATH 522–01 Problem Set #1 solutions 4. Address the two following questions about subrings: √ √ (a) Let Q[ 2, 3] denote √ the smallest subring of R which √ contains√every√rational number, the irrational number 2, and the irrational number 3. Let Q[ 2 + 3] denote the √ small√ est subring of R which contains every rational number and the irrational number 2 + 3 (note √ that √ the previous √ √ question guarantees that these structures are well-defined). Prove that Q[ 2, 3] = Q[ 2 + 3]. √ √ √ √ √ √ It √ suffices to show that 2 and 3 are elements of Q[ 2 + 3] to prove that Q[ 2, √ √ √ √ √ √ √ √ 3] √⊆ Q[ 2 + 3]; likewise, if we show that 2 + 3 ∈ Q[ 2, 3], then Q[ 2 + 3] ⊆ Q[ 2, 3], collectively showing the desired equality. √ √ √ √ √ √ The second of these is trivial: 2 ∈ Q[ 2, 3] and 3 ∈ Q[ 2, 3], so by closure of addition, √ √ √ √ 2 + 3 ∈ Q[ 2, 3]. √ √ √ √ The first membership, however, is more difficult. Since ( 2 + 3) ∈ Q[ 2 + √ √ √ 3], we may use closure of multiplication to find that its square 5 + 2 6 is also in Q[ 2 + 3]. −5 and 12 are √ √ √ 6 Q[ 2 + rational, and can be added and multiplied in turn to the above to find that ∈ √ √ √ √ √ √ √3]. Finally, we multiply 6 by ( 2 +√ 3)√once more to find that√2 3 + 3 √2 is in Q[ 2 + 3], and twice and thrice ( 2 + 3) demonstrates that 2 and − 3 respectively are in √subtracting √ Q[ 2 + 3]. √ √ √ √ (b) Let Z[ 2, 3] and Z[ 2 + 3] be defined as in the previous √ part√except with number” √ “rational √ replaced by “integer” in the definition. Demonstrate that Z[ 2, 3] ̸= Z[ 2 + 3]. √ √ √ √ The same argument as above can be used to show that Z[ 2 + 3] ⊆ Z[ 2, 3]; however, our argument for the reverse containment will not work as designed because we cannot multiply by 21 . It is, however, easy to follow the above argument without the division to guarantee that √ √ √ √ √ 2 2, 2 3, and 2 6 are all elements of Z[ 2 + 3]. However, half of each of these values is not in the ring: note that the set √ √ √ √ {a + 2b 2 + 2c 6 + d( 2 + 3) : a, b, c, d ∈ Z} √ √ √ √ is a ring containing Z and 2 + 3 but not 2 or 3. 5. Prove that if the additive group of a ring R is cyclic, then R is commutative. Let g be a generator of the additive group. Thus, for any a, b ∈ R, either a or −a may be written as g + g + · · · + g for some nonnegative integer k; likewise b or −b may be written as g + g + · · · + g | {z } | {z } k times ℓ times for some non-negative integer ℓ. Using the distributive law, the product ab can be written as ± g + g + · · · + g ; however, likewise ba could be written as ± g + g + · · · + g ; since k and ℓ are ordi| {z } | {z } kℓ times ℓk times nary integers, the number of addends is the same in both expressions, so ab = ba. Page 2 of 2 due Thursday, January 22, 2015